The prognosis of patients with relapsed acute myeloid leukemia (AML) is extremely poor. Allogeneic stem cell transplantation is an effective treatment for AML in first remission, with approximately 60% achieving long-term progression-free survival. However, only 10-20% of patients with relapsed leukemia achieve durable remission. Likewise, patients with advanced remissions continue to have a relatively poor prognosis with stem cell transplantation and improved treatments are required. Our studies have demonstrated that microenvironment/leukemia interactions play a major role in chemoresistance of leukemic stem cells residing in the bone marrow niches. Our preliminary data indicate a key role for SDF-11/CXCR4 interactions in microenvironment-mediated chemoresistance. Inhibition of CXCR4 with the analogue of the only clinically available CXCR4 inhibitor AMD3100 results in mobilization of leukemic cells into the circulation and sensitization to chemotherapy, in part via blockade of the pro-survival signaling pathways. AMD3100 has been extensively used, alone or in combination with G-CSF, for the mobilization of normal hematopoietic stem cells into the circulation. Analysis of apheresis material from AML patients in remission mobilized with AMD3100/G- CSF demonstrated massive egress of AML cells into the peripheral blood, suggesting preferential mobilization of AML blasts. Based on these findings, we propose to test the hypothesis that mobilization of leukemic stem cells by CXCR4 inhibition and G-CSF will result in improved anti-leukemia activity of a standard preparative regimen followed by allogeneic stem cell transplantation. CXCR4 inhibitor AMD3100, G-CSF and busulfan/fludarabine chemotherapy will be administered sequentially to patients with advanced myeloid leukemias in the setting of allogeneic stem cell transplantation. This approach will also avoid potential concerns regarding the mobilization of normal stem cells and associated prolonged life-threatening cytopenias. To test this hypothesis, we propose the following Specific Aims: #1) To conduct a prospective phase I/II clinical trial to determine if addition of AMD3100 (plerixafor), a CXCR4 inhibitor, and G-CSF will improve the antileukemia effect of high dose busulfan-fludarabine conditioning regimen with allogeneic stem cell transplantation for the treatment of AML/MDS. G-CSF will be administered at a standard dose beginning on day -9 daily for 6 days;and AMD3100 will be given 8 h prior to Fludarabine and Busulfan starting on Day -7 for 4 doses. In Phase 1, a maximum of 24 patients will be treated in cohorts of three to determine the safe dose among the three doses 80, 160, 240 mcg/kg of AMD3100. Although the proposed treatment is likely to be safer in patients with low bulk disease because of a lower potential for leukostasis, we will start enrollment in Phase I with patients not in CR, to assure the safety of this approach. In Phase 2, a maximum of 32 additional patients, up to 16 per subgroup (in complete remission (CR) or not in remission (NCR)), will be treated at the AMD3100 dose chosen in Phase 1. The objective of Phase 2 is to determine if the combination of AMD3100 and G-CSF with busulfan, fludarabine will improve progression-free survival post allogeneic stem cell transplantation from an HLA-compatible donor compared to historical controls receiving busulfan-fludarabine alone. #2) to study the in vivo biological effects of disruption of the SDF-11/CXCR4 interactions by AMD3100. These correlative studies will evaluate the relative proportions of mobilized AML and/or normal progenitor cells, analyze expression of the target receptor and adhesion molecules, determine changes in cell cycle distribution of mobilized cells, and assess induction of apoptosis in AML vs normal cells. We anticipate that focus on stroma leukemia crosstalk may result in the development of strategies that alleviate the acquisition of a chemoresistant phenotype and enhance the efficacy of therapies in hematological malignancies and other tumor types. Further, differential analysis of mobilized leukemic and normal cells will provide crucial information for the incorporation of this approach in the up-front therapy for patients with untreated AML or in other hematologic malignancies. PUBLIC HEALTH RELEVANCE: The prognosis of AML patients upon relapse remains poor. AMD3100 and G-CSF are used for stem cell mobilization. In this study we will utilize AMD3100 / G-CSF combination to mobilize leukemic cells from bone marrow niches with the goal to enhance efficacy of busulfan/fludarabine conditioning regimen. This will be done in a setting of allogeneic hematopoietic stem cell transplantation in which hematologic recovery is derived from donor cells.